| Background and ObjectionThe intestinal epithelial surface is composed of a layer of columnar epithelial cells which forms a leaky barrier that directly interfaces with the vast majority of intraluminal bacteria and plays a central role in preventing bacterial translocation. When the barrier is disrupted, the resulting increase in mucosal permeability allows bacteria and toxins to pass from the gut lumen into the submucosa, triggering development of inflammation, endoxemia and even SIRS or MODS, and the inflammation processes further cause disequilibrium in the microenvironment of the barrier and altered gut microbiota.The apical surface of intestinal cells is normally joined by tight junctions (TJ), which form junctional complexes that regulate the integrity and permeability of epithelial monolayers. The properties of TJ are determined by integral membrane proteins, among which occludin and claudins are major components that constitute the backbone of TJ strands. Occludin and claudins both interact with the central scaffolding protein ZO-1, which is the first found member of the zonula occludens family (ZO) localized to TJ and regulating TJ. It has been shown that the TJ-associated network regulates the intestinal mucosal barrier. Disruption of TJ alters intestinal permeability, and dynamic regulation of TJ function is fundamental to many physiological processes. However, the genetic factors implicated in this phenomenon have not been fully revealed yet. A recent study showed that heparan sulfate (HS) proteins, predominantly Sdcl, affect the intestinal barrier, and infusion of HS analogues ameliorates the intestinal barrier defect in patients with protein-losing enteropathy. This suggests that Sdcl may be also involved in the TJ-associated barrier regulation, but the relationship is not well investigated.Syndecan-1 (CD 138, Sdcl), a member of the heparan sulfate proteoglycan family (HSPG), is predominantly expressed on the basolateral surface of epithelial cells, and plays an important role in maintaining cell morphogenesis, promoting tissue repair, establishing cell-cell adhesions and regulating immune function. Several reports have suggested a potential link between Sdcl and regulation of mucosal barrier. Loss of intestinal epithelial Sdcl has been shown to lead to protein leakage in a model of protein-losing enteropathy and abrogate the gut-protective effects of glutamine in a model of gut ischemia-reperfusion. HS moiety of Sdcl could fill the defect of TJ which occurs during breakdown of the intestinal epithelial barrier. However, the precise mechanism by which Sdc1 regulates this barrier function is unknown.In the current study, we hypothesized that Sdcl may participate in TJ-associated network to maintain intestinal epithelial barrier function and prevent bacterial translocation. The aim was to evaluate the effects of Sdc1 on intestinal TJ and epithelial barrier function in vitro and in vivo. These findings may increase the understanding of barrier dysfunction and its initiating events.Materials and results1. Low expression of Sdc1 results in weak barrier function in epithelial cellsWe first examined the Sdc1 expression in four colonic epithelial cell lines. Western blot and real time PCR analysis showed that both HT29 and SW480 had prominent Sdc1 expression, while LoVo and Caco-2 had much lower Sdc1 expression. Because the expression of Sdc1 was high in HT29, but low in Caco-2, these two cell lines were selected as models. Caco-2 cells with stable overexpression of Sdcl were added to the transwell filters and grown to reach confluence for 15 d. HT29 cells transfected with siRNAs were grown at a subconfluent density, and then added to transwell filters at higher density to decrease the time to attain confluence and epithelial differentiation as previously described. EHEC through HT29 was consistently and significantly lower than through Caco-2 cells since 2 h after incubation. Translocation of the EHEC through the barrier was associated with loss of transepithelial electrical resistance (TEER). The FITC-dextran (FD4) flux from the apical to basolateral sides of the monolayers was evaluated without bacterial stimulation. Higher FD4 permeation in Caco-2 monolayers was observed compared with HT29 monolayers, consistent with the higher TEER drop and more bacterial translocation through Caco-2 cells. The EHEC induced cellular toxicity and apoptosis were also detected. At 5 h after infection, the time the maximum TEER response was observed, there was no significantly increased LDH release in culture medium of infected cells over control values. Moreover, no significant caspase-3 cleavage was observed. We also tested the TJ reorganization by immunofluorescent staining. In the control cells, ZO-1 presented a continuous band encircling the cellular junctions. But 5 hours’ infection caused a pronounced decrease in staining intensity and marked discontinuity localized to the intercellular junctions. These data indicate that EHEC induced barrier dysfunction and bacterial translocation are caused by TJ defect, but not due to an infection-induced cell death. Taken together, these results indicate that Sdcl might participate in the maintainence of intestinal barrier function, and the differences in TEER, bacterial translocation and FD4 flux were not due to an infection-induced cell death.2. Sdcl enhances barrier function across epithelial monolayersWhen Sdcl expression was decreased, HT29 cells exhibited significantly decreased TEER and dramatically increased EHEC translocation. The impaired cell layer integrity also caused increased paracellular transport of FD4. The measured fluorescence intensity of FD4 basolaterally in siRNA treated cells was significantly greater compared with control cells. In contrast, after Sdcl overexpression, Caco-2 cells had significantly lower TEER reduction, EHEC translocation and FD4 flux.Stimulation with S. aureus caused a significant drop of TEER in Sdcl siRNA transfected HT29 cells, and also greater S. aureus translocation. The TEER reduction and S. aureus translocation were both lower in Sdcl overexpressed Caco-2 cells.3. Sdcl overexpression restores TJ function during EHEC incubationWhen ZO-1 expression was effectively repressed by siRNA, transfected HT29 cells exhibited remarkable TEER reduction, EHEC translocation and FD4 flux. With increased Sdcl expression in ZO-1 deficient cells, these effects were all significantly decreased. However, the protection of Sdcl overexpression in these cells was still weaker than in cells without ZO-1 deficiency. Otherwise, the results showed the same effects when transfecting occludin siRNA to HT29 cells. These results showed that ZO-1 and occludin are responsible for Sdcl-controlled barrier regulation. Sdcl could restore the dysfunction of ZO-1 and occludin in cell monolayers, and they might cooperate to maintain the barrier function.4. Sdcl co-localizes with ZO-1 and occludin and their expressions change in paralleleIn HT29 cells which had high levels of both Sdcl and TJ proteins ZO-1 and occludin, immunofluorescence imaging showed these proteins co-localized to the cell borders. Western blot analyses showed altered expression of Sdcl was accompanied by changes of TJ proteins. In Caco-2 cells, enhanced Sdcl significantly induced expressions of ZO-1 and occludin compared with control. Relatively, downregulation of Sdcl in HT29 cells led to moderate decrease of ZO-1 and occludin levels.5. Sdcl induces ZO-1 and occludin expressions through activating Stat3 signaling pathwayMoreover, Stat3 was phosphorylated and activated in cells with higher expression of Sdcl. Activation of Stat3 was next induced persistently by infecting HT29 cells with lentiviral construct harboring a constitutively active mutant of Stat3, or temporarily by stimulating cells with IL-6, and induction of protein and mRNA expression of ZO-1 and occludin by Stat3 was also observed. When cells were infected with construct expressing a dominant-negative (DN) Stat3 to block Stat3 binding function, expressions of ZO-1 and occludin slightly decreased.6. Stat3 activity change affects Sdc1-controlled epithelial barrier regulationEctopic expression of Sdc1 caused increased TEER but decreased EHEC translocation and FD4 flux compared with its control. When Stat3 activity was suppressed, increases in ZO-1 and occludin during Sdcl overexpression were eliminated. Consistently, the Sdc1 maintained barrier was impaired in cells treated with Stat3 DN lentivirus in combination with Sdc1 lentivirus, compared with cells treated with Sdcl lentivirus alone, shown by increased TEER reduction, EHEC translocation and FD4 flux.7. Stat3 binds to promoter regions of ZO-1 and occludin directlyApproximately 10 kb of genomic DNA sequence from the human genomes were queried for predicted Stat3 binding sites. The Stat3 binding sites were observed more than 10 times. After Stat3 was activated, its binding was remarkably enriched in the promoter regions of ZO-1 and occludin. Moreover, Stat3 binding to the ZO-1 and occludin promoters was increased in Caco-2 cells with Sdcl overexpression and decreased in HT29 cells with Sdcl knockdown.8. ZO-1 and occludin interacts with each otherIn Co-IP assays using HT29 cells, ZO-1-specific immune complex contained endogenous occludin, and vice versa. Neither ZO-1 nor occludin could be immunoprecipitated by control IgG. Furthermore, western blot analysis showed when ZO-1 or occludin protein level was decreased, Sdcl expression of the transfected cells increased.9. Sdcl prevents intestinal bacterial translocation in mouse colitis modelMesenteric lymph nodes (MLN), liver and spleen tissues from a DSS-induced mouse colitis model were analyzed for signs of bacterial translocation. No bacteria were detected in any tissues of mice in the PBS control group. However, quantification of bacterial content in tissues of mice in the DSS control group revealed greater bacterial invasion compared to Sdcl-pretreated and Sdcl-posttreated groups, and the numbers of Enterococcus and Enterobacteriaceae detected in spleens were especially higher than other strains. In addition, the numbers of bacteria were less in livers and spleens. For a number of bacteria including Bifidobacterium, Bacteroides, Enterococcus and Clostridium, the average numbers detected in MLNs were greater in the Sdcl-posttreated group, compared to those in the Sdcl-pretreated group, suggesting a greater efficacy of Sdcl before DSS challenge.10. Sdcl restores ZO-1 and occludin expressions and barrier function in miceWe assessed whether supplementation of Sdcl enhanced ZO-1 and occludin expressions in vivo using the same DSS-induced Sdcl-loss mouse model. Western blots of Sdcl in the intestinal mucosa showed the DNA transfection in mice increased Sdcl levels and activated Stat3 pathway in Sdcl-pretreated or posttreated group compared to the control groups. In the DSS control group, Sdcl was decreased in colon of mice, and ZO-1 and occludin proteins also disappeared. Higher levels of ZO-1 and occludin were found in the Sdcl-pretreated and posttreated group. Colons in the DSS control group showed destruction of the epithelial architecture, loss of epithelial integrity, mucus gland damage, some multifocal shallow ulcers, and intense inflammatory cell infiltration in lamina propria. However, these lesions were less prominent in the Sdcl-pretreated or posttreated group. In the DSS group, but not in the control, diarrhea appeared on day 4 and bloody stool occurred on day 5. Weight loss and shortened colons could also be observed. While in the Sdcl-pretreated and posttreated groups, these colitis symptoms were not evident. Consistently, DAIs were significantly lower in the mice treated with Sdcl than in the control mice. In the PBS control group, the TJ complex was situated close to the lumen and appeared as typical membrane fusions. There was an extreme narrowing of the intercellular gap, indicating an intact intestinal mucosal barrier. In the colonic epithelium of the DSS control and vector-pretreated or posttreated group, the junctions were smaller, loose and indistinct, and the intracellular space was enlarged. However, colons from mice in the Sdcl-pretreated or posttreated group showed obvious TJ which remained tight, suggesting that Sdcl could also attenuate DSS-induced morphological injury.ConclusionThe current data indicate that TJ-associated Sdcl is involved in the maintainence and regulation of the intestinal mucosal barrier. Supplementation of exogenous Sdcl enhances ZO-1 and occludin expressions, so as to act in synergy with TJ, consolidate barrier function, and inhibit bacterial translocation, especially when given before barrier dysfunction occurs. These findings present potential therapeutic targets for patients suffering diseases due to gut mucosal barrier dysfunction and bacterial translocation. |
PDF Full Text Request